Anthropic’s massive IPO comes with staggering infrastructure commitments and steep operating losses.
Anthropic is expected to price its initial public offering on Wednesday, Nov. 18, ahead of a planned Nasdaq debut on Nov. 19, in what is shaping up to be the largest artificial intelligence IPO to date.
The company launched its roadshow on Tuesday, marketing 825 million shares at $70 to $78 each, including 801.4 million newly issued shares and 23.6 million from existing shareholders. The offering could raise as much as $64.35 billion and, based on the shares outstanding disclosed in its filing, value Anthropic at approximately $1 trillion at the top of the range.
The proposed valuation is below the $1.5 trillion figure that circulated before a recent selloff in technology stocks, suggesting the company is prioritizing a successful debut and strong investor demand over pursuing the highest possible valuation.
Pricing at roughly $1 trillion, with the option to move toward $1.2 trillion, “is the right move,” according to a Nov. 10 research note from merchant bank Yorkville Ives & Co. “A deal of this size gets one shot at its debut, and making it digestible for the market matters more than the last turn of valuation,” the firm added.
An Unusual Sidepot
Alongside the public offering, Anthropic is arranging a $26 billion private placement with seven major tech investors: Alphabet Inc., Nvidia Corp., Advanced Micro Devices Inc., Broadcom Inc., Micron Technology Inc., Microsoft Corp., and Taiwan Semiconductor Manufacturing Co. The combined fundraising could reach $90.4 billion.
The arrangement gives companies with significant commercial relationships with Anthropic — including chip suppliers, cloud providers and distribution partners — stakes in the “Claude” developer before its first public trading session. It’s also a unique setup since several of these investors could also make money from Anthropic’s infrastructure spending.
| ANTHROPIC SNAPSHOT |
|---|
| Index: NASDAQ |
| Ticker Symbol: ANT |
| Offering: 825M shares |
| Price: $70-$78 per share |
| Proceeds: $64.35B |
| Q3 Revenue: $16.96B |
| Operating Loss: $567M |
| Infrastructure Commitments: $900.9B |
| Valuation: $1 Trillion |
| Private Placement: $26B |
“Google, Nvidia, AMD, Broadcom, Micron, Microsoft and TSMC have commercial interests in the technology Anthropic needs to expand, giving them potential benefits beyond the return on their shares,” Varun Datta, venture capitalist and Founder of Truth Ventures, told Global Finance. “What’s unusual is having seven major technology companies investing at this scale, including businesses that compete with one another and could also supply Anthropic with the infrastructure it needs.”
Google and Microsoft sell cloud computing capacity, while Nvidia, AMD and Broadcom supply AI chips and related technology. Their potential returns extend beyond Anthropic’s share price because its continued expansion could generate billions in additional orders for their businesses, he explained.
“I’d want to examine whether the investments are tied to purchasing commitments, cloud contracts or preferential supply arrangements,” Datta added.
“Those details matter because they would help establish how much of the capital raised is available for Anthropic to deploy independently, and how much could ultimately flow back to its investors through commercial spending.”
Amended Terms
The proposed range of $70 to $78 a share on 825 million shares tops out below the terms in a filing Anthropic briefly sent to banks — $70 to $80 a share on 866.7 million shares — which a company representative later said contained inaccurate pricing details, Reuters reported.
Anthropic did not respond to Global Finance’s request for a comment.
The startup’s revised terms follow a sharp tech selloff that began Nov. 5, when the Nasdaq fell 5.5% — its worst day since April 2025 — amid concerns that Democrats’ midterm gains could bring tighter AI regulation.
Michigan Senator-elect Abdul El-Sayed, who won on Nov. 3, campaigned on a plan that includes “pausing frontier AI development,” the category that covers Anthropic, according to Harvard’s Ash Center.
A Closely Intertwined System
Google, Amazon.com Inc. and Microsoft are among Anthropic’s backers, customers and computing suppliers, illustrating how tight-knit the AI ecosystem has become.
The company’s substantial infrastructure commitments offer a multi-year demand signal for chipmakers and cloud providers, while raising questions about whether revenue growth can eventually translate into sustainable margins.
The central question for investors will be whether Anthropic can maintain its rapid expansion while bringing computing costs under control. Its second-quarter operating profit and third-quarter loss offer sharply different pictures of the company’s underlying earnings potential.
The offering also comes on the heels of SpaceX’s massive IPO and OpenAI’s expected public listing in 2027. But while Anthropic’s market debut could establish an important benchmark for valuing privately held AI companies, the longer-term test will be whether it can sustain that valuation after the initial excitement fades.
A sharp decline in Anthropic’s share price, as seen with SpaceX, which is down more than 13% year to date, could challenge the investment assumptions underpinning the broader AI investment boom.
Anthony Noto covers corporate finance and private credit. Contact him at anoto@gfmag.com.